R/authorships.R

In genderizeR: Gender Prediction Based on First Names

#' Authorships sample
#'
#' A dataset containing a simple random sample of authorships 
#' (unique combination of authors and titles) from WebOfScience records 
#' of articles of "biographical-items" or "items-about-individual" types
#' from all fields of study
#' published from 1945 to 2014.
#' The sample was drawn in December 2014. 

#'
#' @format A data frame with 2641 rows and 5 variables:
#' \describe{
#'   \item{title}{The title of an article.}
#'   \item{authors}{All the authors of the article.}
#'   \item{value}{A single author of the article - with the title forms an authorship; there can be several authorships per article.}
#'   \item{genderCoded}{Manually coded gender of an author.
#'    There are four codes: "female", "male", "noname", "unknown". 
#'    "Noname" is the code for a case were human coders were not able to find 
#'    a first name of an author. "Unknown" is the code for a case were 
#'    the coders found a full name of an author but were not able to verify 
#'    if she or he is a man or a female.}
#'   \item{WOSaccessionNumber}{The original ID of an article 
#'   in WebOfScience database.}
#' }
#' @source \url{http://webofknowledge.com/}
#' 
#' 

"authorships"

# codedAuthorships = readRDS("data-raw/codedAuthorships.rds")
# head(codedAuthorships)
# library(dplyr) 
# authorships = codedAuthorships %>% select(WOSaccessionNumber, title, authors, value, genderCoded)
# devtools::use_data(authorships, overwrite = TRUE)